[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Fwd: Re: [Platform-testers] new snapshot available: coreutils-8.12.1
From: |
Markus Duft |
Subject: |
Re: Fwd: Re: [Platform-testers] new snapshot available: coreutils-8.12.178-df9cd |
Date: |
Thu, 08 Sep 2011 14:49:51 +0200 |
User-agent: |
Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9.2.17) Gecko/20110706 Lightning/1.0b3pre Thunderbird/3.1.10 |
On 09/08/11 14:00, Jim Meyering wrote:
> Markus Duft wrote:
>> On 09/08/11 10:17, Jim Meyering wrote:
>>> Markus Duft wrote:
>>>
[snip]
>
> Thanks. I've pointed out a few issues, hoping you'll adjust
> accordingly and resubmit.
sure :) this exercised my git-rebase-foo a lot though ...
>
>> setgroups is not needed here, as it something different. i have 2
>
> I added that conjunct quite deliberately.
> If we make the compromise of not detecting the getgr* performance
> problems, then we need some other way to ensure that we use the
> _nomembers functions only when needed. Appending " && ! HAVE_SETGROUPS"
> was my way of saying to use this kludge only on a system that also has
> the defect of a missing setgroups function. Yes, that deserves a
> comment.
hm. ok. i added and commented this, although it feels quite wrong for me. is
there any platform that has the _nomembers functions? if yes, would it do any
harm to use them? idk...
>
> ...
[snip]
>
> Please don't add it here, since it is unrelated to this ancient SCO hack.
> Add it on a separate line (in jm-macros.m4), with its own comment.
heh. while looking for a location to add the check, i realized there is one
already! :) so i ommitted this step. hope thats ok, my build goes through, but
for obvious reasons this is not significant (as i'll never have it defined). a
quick grep over the build tree yields this:
mduft coreutils-8.12.193-d8dc8 $ find . -type f | xargs grep SETGROUPS
./autom4te.cache/traces.0:m4trace:configure.ac:46: -1-
AH_OUTPUT([HAVE_SETGROUPS], [/* Define to 1 if you have the `setgroups\'
function. */
./autom4te.cache/traces.0:@%:@undef HAVE_SETGROUPS])
./ChangeLog-2006: * src/setuidgid.c (main) [! HAVE_SETGROUPS]: Don't call
setgroups.
./lib/config.h:/* #undef HAVE_SETGROUPS */
./lib/config.hin:#undef HAVE_SETGROUPS
./lib/config.hin~:#undef HAVE_SETGROUPS
./src/chroot.c:#if ! HAVE_SETGROUPS
./src/setuidgid.c:#if HAVE_SETGROUPS
./src/system.h: other than interix, check for HAVE_SETGROUPS, as interix is
./src/system.h:#if ! HAVE_SETGROUPS
>
[snip]
hope the new versions are acceptable. otherwise, i'll be happy to adjust once
more, and resubmit tomorrow :)
regards,
markus
0001-build-accommodate-missing-setgroups-on-Interix.patch
Description: Text Data
0002-build-use-getgr-_nomembers-functions-on-interix.patch
Description: Text Data
- coreutils-8.12.178-df9cd on Solaris 8, (continued)
- coreutils-8.12.178-df9cd on Solaris 8, Bruno Haible, 2011/09/01
- coreutils-8.12.178-df9cd on Cygwin 1.5, Bruno Haible, 2011/09/01
- coreutils-8.12.178-df9cd on mingw, Bruno Haible, 2011/09/01
- coreutils-8.12.178-df9cd on Cygwin 1.7.5, Bruno Haible, 2011/09/01
- Re: Fwd: Re: [Platform-testers] new snapshot available: coreutils-8.12.178-df9cd, Jim Meyering, 2011/09/07
- Re: Fwd: Re: [Platform-testers] new snapshot available: coreutils-8.12.178-df9cd, Markus Duft, 2011/09/08
- Re: Fwd: Re: [Platform-testers] new snapshot available: coreutils-8.12.178-df9cd, Jim Meyering, 2011/09/08
- Re: Fwd: Re: [Platform-testers] new snapshot available: coreutils-8.12.178-df9cd, Markus Duft, 2011/09/08
- Re: Fwd: Re: [Platform-testers] new snapshot available: coreutils-8.12.178-df9cd, Jim Meyering, 2011/09/08
- Re: Fwd: Re: [Platform-testers] new snapshot available: coreutils-8.12.178-df9cd,
Markus Duft <=
- Re: Fwd: Re: [Platform-testers] new snapshot available: coreutils-8.12.178-df9cd, Jim Meyering, 2011/09/08
- Re: Fwd: Re: [Platform-testers] new snapshot available: coreutils-8.12.178-df9cd, Markus Duft, 2011/09/12
- Re: Fwd: Re: [Platform-testers] new snapshot available: coreutils-8.12.178-df9cd, Jim Meyering, 2011/09/12